You are on page 1of 71

1

Fundamentals of Earthquake Engineering


by

Amr S. Elnashai and Luigi Di Sarno

Context, Framework and Scope

Earthquakes are one of the most devastating natural hazards that cause great loss of life and livelihood. On average, 10,000 people die each year due to earthquakes, while annual economic losses are in the billions of dollars and often constitute a large percentage of the gross national product of the country affected. Over the past few decades earthquake engineering has developed as a branch of engineering concerned with the estimation of earthquake consequences and the mitigation of these consequences. It has become an interdisciplinary subject involving seismologists, structural and geotechnical engineers, architects, urban planners, information technologists and social scientists.

Context, Framework and Scope (continued)

This book frames earthquake structural engineering as a balance between Demand or imposed actions and deformations, and Supply or capacity to withstand actions and deformations. The proposed systems framework is referred to as From Source-to-Society.

Chapter 1 Earthquake Characteristics

Earthquake Characteristics
Causes of earthquakes
Plate tectonics theory Faulting Seismic waves

Measuring earthquakes
Intensity Magnitude Intensity-magnitude relationships

Source-to-Site-Effects
Directional Effects Site Effects Dispersion and Incoherence

Effects of earthquakes
Damage to buildings and lifelines Effects on the ground Human and financial losses

Plate Tectonics Theory


Earthquake causes

An earthquake is manifested as ground shaking caused by the sudden release of energy in the Earths crust. Earthquake occurrence may be explained by the theory of large-scale tectonic processes referred to as plate tectonics. Intense seismic activity occurs predominantly on known plate boundaries.

Tectonic plates (left) and worldwide earthquake distribution (right)

Plate Tectonics Theory (continued)


Plate tectonics
Plates are large and stable rigid rock slabs with a thickness of about 100 km forming the crust or lithosphere and part of the upper mantle of the Earth.

Large tectonic forces take place at the plate edges due to the relative movement of the lithosphere-asthenosphere complex.

Cross-section of the Earth with the main type plate boundaries

Plate Tectonics Theory (continued)

The principal types of plate boundaries


Divergent or rift zones: plates separate themselves from one another and either an effusion of magma occurs or the lithosphere diverges from the interior of the Earth. Convergent or subduction zones: adjacent plates converge and collide. A subduction process carries the slab-like plate, known as the under-thrusting plate into a dipping zone. Transform faults or transcurrent horizontal slip: two plates glide past one another but without creating new lithosphere or subducting old lithosphere.

Plate Tectonics Theory (continued)


Divergent plate boundaries form narrow bands of shallow earthquakes at midoceanic ridges. Shallow and intermediate earthquakes also occur at convergent zones. Shallow earthquakes with large magnitudes may occur along transform faults. Deep earthquakes are generally located in subduction zones.

Tectonic mechanisms at plate boundaries

Plate Tectonics Theory (continued)


Classification of tectonic earthquakes
Inter-plate earthquakes

10

Plate boundary earthquakes which contribute 95% of world-wide seismic energy release.

Intra-plate earthquakes
The genesis of this seismic activity is attributed either to the geological structural complexity of the lithosphere or anomalies in its temperature and strength. Intra-plate earthquakes generally fall in two groups: - Plate boundary-related earthquakes; - Mid-plate earthquakes.
Earthquake (type) Inter-plate Intra-plate (plate boundary related) Intra-plate (mid-plate) Slip rate (v) (mm/year) v > 10 0.1 v 10 v < 0.1 Recurrence time (year) ~ 100 102 ~ 104 > 104

Faulting
Faults

11

When two ground masses move with respect to one another, elastic strain energy due to tectonic processes is stored and then released through the rupture of the interface zone. The distorted blocks snap back towards equilibrium and earthquake ground shaking is produced. This process is referred to as elastic rebound. The resulting fracture in the Earths crust is termed a fault.

Active faults: Carrizo Plains (San Andreas fault) in California (left) and the Corinth Canal in Greece (right)

Faulting (continued)
Parameters used to describe fault dimensions and motion

12

Azimuth (): the angle between the trace of the fault (defined as the intersection of the fault plane with the horizontal) and the northerly direction; Dip (): the angle between the fault and the horizontal plane; Slip or rake (): the angle between the direction of relative displacement and the horizontal direction; Relative displacement (u): the distance travelled by a point on either side of the fault plane; Area (S): surface area of the highly stressed region within the fault plane.
Z (Zenith) X (North) Horizontal plane

Fault trace 180-

u Fault plane S

Parameters used to describe fault dimension and motion

Faulting (continued)
Fundamental fault mechanisms
Dip-slip faults: one block moves vertically with respect to the other. Strike-slip faults: the adjacent blocks move horizontally past one other.

13

Fault plane

Fault plane

Fault trace

Fault plane

F F
F

Dip ()

Normal Fault
Reverse Fault Fault plane
F F
F F

Foot wall Hanging wall

Fault plane Left-Lateral Strike Slip Fault


Right-Lateral Strike Slip Fault

Fundamental fault mechanisms

Faulting (continued)
Fault mechanisms
North rake () foot wall dip () +180 -180 hanging wall normal + 0 right azimuth (strike, )

14

Fault mechanisms as a function of the rake ()


reverse fault right slip reverse fault left slip

reverse

left

dip () 0< <90 (oblique fault)


pure reverse fault (= 90) pure normal fault (= -90)

normal fault right slip

normal fault left slip

= 90 (vertical fault)

Pure strike slip Pure strike slip (left lateral, =0) (right lateral, =180)

down slip (= 90)

Faulting (continued)
Source parameters

15

The focus or hypocentre of an earthquake is the point under the surface where the rupture is said to have originated. The projection of the focus on the surface is termed epicentre. Earthquake foci are located by geographical coordinates, namely latitude and longitude, the focal depth and the origin or occurrence time.

Definition of source parameters

Faulting (continued)

16

Source parameters
Many earthquakes have focal depths in the range of 5-15 km, while intermediate events have foci at about 20-50 km and deep earthquakes occur at 300-700 km under the surface. The three types are also referred to as shallow, intermediate and deep focus, respectively. Crustal earthquakes normally have depths of about 30 km or less. The source is not a single point, hence the distance from the source required for engineering seismology applications is ill-defined.

Problem 1.1

17

Determine the source mechanism of an earthquake with a fault that has a dip = 60 and a rake = 45. Comment on the results.

Seismic Waves

18

Earthquake shaking is generated by two types of elastic seismic waves: body and surface waves.

Body waves
They include longitudinal or primary waves (also known as P-waves) and transverse or secondary waves (also referred to as S-waves). Body waves attenuate in inverse proportion to the distance x.

Surface waves
Surface waves include Love (indicated as L- or LQ-waves) and Rayleigh (indicated as R- or LR-waves) waves. Surface waves attenuate in inverse proportion to the square root of the distance.

Seismic Waves (continued)


P-waves
P-waves cause alternate compression and dilation in the rock.

19

S-waves
S-wave propagation causes vertical motion and horizontal side-to-side motion.

Small amplitudes and short periods. P-waves travel faster, at speeds between 1.5 and 8 km/sec in the Earths crust.

Large amplitudes and long periods. S-waves, are slower, usually have 50% to 60% of the speed of P-waves.

Seismic Waves (continued)


The difference in velocities of P- and S- waves may be employed to locate the earthquake origin.

20

Assuming homogenous soil profiles between earthquake foci and observation sites, the focal distance x is linearly dependent on the time-lag t between the arrival of P- and S-waves.
Layer (type) Crust Upper Mantle Lower Mantle Outer Core Inner Core Depth (km) 10 - 30 40 220 400 670 1200 2885 2890 3800 5150 5155 6371 vp (km/s) 6.57 8.12 8.06 9.13 10.75 11.78 13.72 8.06 9.31 10.36 11.03 11.26 vs (km/s) 3.82 4.42 4.35 5.22 5.95 6.52 7.26 0.00 0.00 0.00 3.50 3.67

v P vS x = t v P vS
18.0 16.0 Ratios (vPvS) / (vP-vS) 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 10-30
Crust

(1.1 a)

Proposed step-function

Omori's formula

For a quick evaluation, Omoris formula (Kanai, 1983) may also be used (x in km, t in seconds)

40

220

400
Upper Mantle

670

1200

2885

Lower Mantle

Depth of Earth's Layers (in km)

x 7.42 t

(1.1 b)

Seismic Waves (continued)


The procedure to locate an earthquake epicentre and origin time is as follows: a) Obtain seismogram records for a given observation site; b) Select the arrival time of the body waves on the record traces; c) Compute the time delay t in the arrival of P- and S-waves; d) Subtract the time delay (or travel time) t from the arrival time at the observation site to derive the origin time;

21

e) Use eqns. (1.1) to evaluate the distance x between the seismic station and the epicentre; f) Draw a circle on a map around the station location (or centre) with a radius equal to x; g) Repeat steps (a) to (f) for a second seismic station; h) Repeat steps (a) to (f) for a third seismic station.

Seismic Waves (continued)


Love (L- or LQ-) waves
LQ-waves are generated by constructive interference of SH body waves.

22

Rayleigh (R- or LR-) waves


LR-waves are caused by constructive interference of body waves, such as P and SV.

LQ-waves have large amplitudes and long periods. For a layered soil, LQ-wave velocity vLQ is generally: with vS1 and vS2 the velocities of Swaves in the surface and deeper layers, respectively.
v S1 < vLQ < v S2

LR-waves exhibit very large amplitude and regular wave-forms. LR-waves are slower than S-waves; it may be assumed that:

v LR 0.92 v S

Seismic Waves (continued)

23

Body waves are reflected and refracted at interfaces between different layers of rock according to Snells law of refraction from elementary optics. When reflection and refraction occur part of the energy of one type is transformed into the other. Regardless of whether the incident wave is P or S, the reflected and refracted waves, also termed multiple phase waves, each consist of P- and Swaves. Their name indicates the travel path and mode of propagation. For example, SP starts as S and then continues as P. Multiple phase waves do not possess significant damage potential. However, when P- and S-waves reach the ground surface they are reflected back. Such reflection may lead to significant local amplification of the shaking at the surface. It has been shown that seismic waves are influenced by soil conditions and local topography.

Problem 1.2

24

Locate and mark on the map provided the epicentre of an earthquake that was recorded in Italy by three observation sites 1, 2 and 3, with a time delay between P- and S-waves of 5.0, 7.5 and 6.0 seconds, respectively. The body wave velocities are 8.5 km/sec and 4.30 km/sec; it is up to the reader to determine which of these values refer to P- and S-waves. Compare the results obtained by eqn. (1.1-a) with those estimated through eqn. (1.1-b).

Measuring Earthquakes

25

Earthquake size can be expressed in several ways.

Intensity
Qualitative (non-instrumental) measurement. Subjective damage evaluation based on description of earthquake effects.

Magnitude
Quantitative (instrumental) measurement. Objective measure of earthquake size or fault dimensions.

Intensity
Discrete scales are used to quantify seismic intensity. The levels are represented by Roman numerals and each degree of intensity provides a qualitative description of earthquake effects: i. Mercalli_Cancani-Seiberg (MCS): 12-level scale used in southern Europe;

26

ii. Modified Mercalli (MM): 12-level scale used in North American and several other countries; iii. Medvedev-Sponheuer-Karnik (MSK): 12-level scale developed in Central and Eastern Europe and used in several other countries; iv. European Macroseismic Scale (EMS): 12-level scale adopted since 1998 in Europe; v. Japanese Meteorological Agency (JMA): 7-level scale used in Japan.

Intensity (continued)
MODIFIED MERCALLI

27

II

III

IV

VI

VII

VIII

IX

XI

XII

MERCALLI CANCANI SEIBERG

II

III

IV

VI

VII VIII

IX

XI XII

MEDVEDEV SPONHEUER KARNIK

II

III

IV

VI

VII

VII

IX

XI

XII

EUROPEAN MACROSEISMIC

II

III

IV

VI

VII

VIII

IX

XI

XII

JAPANESE METEOROLOGICAL AGENCY

II

III

IV

VI

VII

Comparison and equivalence between seismic intensity scales

Intensity (continued)
Intensity scales are used to plot contour lines of equal intensity or isoseismals.

28

The isoseismal map of the October 17, 1989 Loma Prieta earthquake in California is shown below. It locates the epicentre (marked as a star) and provides Modified Mercalli (MM) intensities between isoseismals (Roman numerals), and MM intensities at specific cities (Arabic numerals).

Intensity (continued)

29

Structures in the immediate vicinity of earthquake sources experience very high ground accelerations but may sustain little or no damage, depending on their periods of vibration. On the other hand, intensity is a measure of the perceptibility of the earthquake and its actual consequential damage. Relating intensity to peak ground acceleration is, in principle, illogical. The necessity of bridging the distance between historical earthquake observations (based mainly on intensity) and code-defined forces (based entirely on peak ground acceleration) warrants the efforts expended in correlating the two measures. The measurement of earthquake size should be based on the amount of energy released at the focus. Therefore, magnitude scales have been defined as presented hereafter.

Magnitude
Local (or Richter) magnitude (ML)

30

It is a measure of the maximum seismic wave amplitude A (in millimetres) recorded on standard Wood-Anderson seismographs located at a distance of 100 kms from the earthquake epicentre. Magnitude ML is related to A by the following relationship:

M L = log (A ) - log(A 0 )
where A0 is a calibration factor that depends on distance. Earthquakes with ML greater than 5.5 cause significant damage, while an earthquake of ML=2 is the smallest event normally felt by people.

Magnitude (continued)
Body wave magnitude (mb)
It measures amplitudes of P-waves with periods of about 1.0 second. This scale is suitable for deep earthquakes. Magnitude mb is related to the amplitude A and period T of P-waves as follows:

31

A m b = log + ( ) T
in which () is a function of the epicentre distance (in degrees).

Surface wave magnitude (MS)


It is a measure of the amplitudes of LR-waves with a period of 20 seconds. MS is used for large earthquakes. The relation between amplitude A, period T, distance and MS is given by:

A M S = log + 1.66 log( ) + 3.30 T


where is in degrees, the ground motion amplitude in microns and the period in seconds.

Magnitude (continued)

32

Moment magnitude (MW)


It accounts for the mechanism of shear that takes place at earthquake sources. It is not related to any seismic wave. Moment magnitude is defined as a function of the seismic moment M0:

M 0 = G A u
in which G is the shear modulus of the material surrounding the fault, A is the fault rupture area and u is the average displacement slip between opposite sides of the fault. Mw is thus given by:

M w = 0.67 log (M 0 ) 10.70

where M0 should be expressed in ergs.

Magnitude (continued)
Magnitude scales do not increase monotonically with earthquake size. Saturation is evident as magnitude increases. Moment magnitude (Mw) is the only magnitude scale which does not suffer from the saturation problem for great earthquakes: M ~ Mw.
9
MS

33

8
mB

MJMA ML mb

7
Magnitude

6 5 4 3 2 2 3 4 5 6 Moment Magnitude Mw 7 8
M~Mw

Saturation of magnitude scales

Magnitude (continued)

34

Type ML mb Ms Mw

Author Richter, 1935 Gutenberg & Richter, 1956 Richter & Gutenberg, 1936 Kanamori, 1977

Earthquake Size Small Small-toMedium Large All

Earthquake Epicentre Depth distance Shallow Deep Shallow All < 600 km > 600 km > 2000 km All

Reference Parameter Wave amplitude Wave amplitude (P-waves) Wave Amplitude (LR-waves) Seismic Moment

Applicability Regional (California) World-wide World-wide World-wide

Saturatio n n.a.

The general correlation between magnitude M and A is as follows:

M = log (A ) + f (d, h ) + C S + C R
in which the function f(d,h) accounts for epicentral distance d and focal depth h. The coefficients CS and CR are station and regional corrections, respectively.

Magnitude (continued)
Earthquake magnitude and seismic energy
Earthquake magnitude can be used to quantify the amount of energy released during fault rupture. Relationships between magnitudes and seismic energy exist, e.g.,

35

log (E ) = 1.5 M S + 11.8


where E is in ergs. As the magnitude increases by one unit, the energy increases by a factor of 31.6. mb and MS are related to seismic energy E as follows:

log (E ) = 2.4 m b 1.3 log (E ) = 1.5 M S + 4.2

where E is expressed in joules.

Magnitude (continued)
Magnitude-moment relationships

36

Seismic moment M0 measures the energy E released by fault rupture during earthquakes. The following relationship is applicable to all source mechanisms:

E=

M0 2G

where is the stress drop = 1-2 and G is the shear modulus of the material surrounding the fault. Magnitude-moment relationships have been defined empirically for periods less than 20 seconds:

log (M 0 ) = 1.5 M S + 16.1


Body wave magnitude mb can be related to MS over a wide range by the following semi-empirical formula proposed by Gutenberg and Richter:

m b = 0.63 M S + 2.5

Magnitude (continued)

37

Correlation between magnitude and energy release

Magnitude (continued)
Earthquake magnitude and fault rupture parameters for inter-plate earthquakes
MS as a function of the fault rupture length L:
mean value: 95th percentile:
0.95 M S = M S (L ) + 0.52

38

M S (L ) = 6.04 + 0.71 log(L )

Surface wave magnitude MS related to the maximum observed displacement of fault D:

M S = a + b log (D )

where the displacement D is in metres, while coefficients a and b are given below.
Fault Mechanism Normal Reverse Strike-slip a 6.67 6.79 6.97 b 0.75 1.31 0.80

Relationship between the fault surface rupture S and surface magnitude MS:

M S = 4.15 + log (S)

Magnitude (continued)
Earthquake magnitude and fault rupture parameters for intra-plate earthquakes
Average source parameters and relevant magnitude scales are listed below:
Rupture length (km) 2.1 3.8 7.0 13.0 24.0 45.0 58.0 75.0 85.0 Slip (m) 0.01 0.03 0.11 0.34 1.10 3.70 5.80 9.20 11.50 mb 4.5 5.0 5.5 6.0 6.5 7.0 7.2 7.4 7.5 MS 3.35 4.35 5.35 6.35 7.35 8.32 8.53 8.87 9.00 log M0 (dyne-cm) 22.2 23.2 24.2 25.2 26.2 27.2 27.6 28.0 28.2

39

Using the equations for inter-plate earthquakes may overestimate fault rupture lengths.

Intensity-Magnitude Relationships

40

Relationship between local magnitude ML and epicentral intensity I0 for Southern California:

M L = 0.67 I 0 + 1.00
Relationship between mb and I0 which is useful in the central and eastern USA

m b = 0.49 I 0 + 1.66
Intensity-magnitude relationships proposed by Ambraseys: For north-west Europe For the Alpine zone in Europe M S = 0.90 + 0.58 I i + 1.10 10 3 ri + 2.11 log ( ri ) where Ii the Modified Mercalli intensity of the ith isoseismal and ri the radius of equivalent area enclosed by the ith isoseismal. Local geological conditions and focal depths can significantly affect the intensity of earthquake ground motion. Semi-empirical formulations:

M S = 1.10 + 0.62 I i + 1.30 10 3 ri + 1.62 log (ri )

M S = 0.66 I 0 + 1.70 log (h ) 1.40


where the focal depth h should be expressed in kilometres and the intensity I0 in the MM scale.

Problem 1.3

41

Calculate the surface wave magnitude MS for an earthquake with IMM of VII, in an area that can be approximated by a circle with radius 20 km for a site at the borders of the given isoseismal map. This site is located in the Western United States. You may use the equation given below. Compare the ensuing value with the estimates from relationships with other magnitude scales. Calculate the fault surface displacements. Assume that the earthquake mechanism is normal.

M S = 1.10 + 0.62 I i + 1.30 10 3 ri + 1.62 log (ri )

Source-to-Site Effects

42

The characteristics of seismic waves are altered as they travel from the source to the site of civil engineering works. Local site conditions may affect significantly the amplitude and frequency content of earthquake ground motions. Nonlinearity of soil response and topographical effects may also significantly influence ground motion parameters. The most important topographical parameter influencing local amplification of ground motion is the steepness of a ridge.

Source-to-Site Effects (continued)

43

Structure

Conditions

Type

Size

Quantitative Predictability

Poor: generally Amplification at top of under-predict size; Ranges up to a factor of Sensitive to shape ratio, structure, rapid may be due to ridgeSurface 30 but generally from largest for ratio between changes in amplitude ridge interaction and topography 0.2 and 0.6 about 2 to 20 phase along slopes three-dimensional effects Good: away from Effects most pronounced Broadband One-dimensional models edges one dimension Shallow and wide amplification near near edges; largely works well, near may under-predict at (depth/width edges due to vertically propagating edges extend one higher frequencies by <0.25) sedimentgeneration of surface shear waves away from dimension to higher about 2 near edges filled valleys waves edges frequencies One-dimensional models Fair: given detailed may under-predict for a Deep and narrow Broadband (depth/width Effects throughout valley amplification across wide bandwidth by about 2 description of vertical >0.25) sediment- width valley due to whole to 4; resonant frequencies and lateral changes in material properties shifted from one filled valleys valley modes dimensional analysis

Effects of topographic and subsurface irregularities

Directional Effects

44

Directivity
Directivity occurs because fault ruptures are moving wave sources which travel at a finite velocity along the fault and beyond. The engineering implication of such directivity effects is that sites which are equidistant from the source will be subjected to varying degrees of shaking from the same earthquake. Rupture directivity also causes the polarisation of ground motion. Polarisation causes more intense shaking in the fault-normal direction than in the fault-parallel direction.

Directional Effects (continued)

45

As the fault rupture moves away from the epicentre it generates ground motion from each segment of the breaking fault. The ground motion radiates outwards in all directions and the seismic energy propagates through expanding wave fronts. The over-riding of stress waves or constructive interference results in larger ground motion magnification with shorter total duration in the direction of rupture propagation.

Directivity effects on sites toward and away from direction of fault rupture

Site Effects
Site resonance effect
The ratio between the period of the site and that of the building is important in estimating the amplification effects; this is referred to as site resonance. The site period TS for uniform single soil layer on bedrock can be estimated from the relationship:

46

TS =

4H vS

where H and vS are the depth of soil layer and soil shear wave velocity, respectively. Periods associated to higher modes can be determined as follows:

TS, n =

1 4H 2 n 1 v S

in which n represents the n-th mode of vibration (n >1).

Site Effects (continued)

47

Material (type) Loose saturated sand Sandy clay Fine saturated sand Clay/sand mix Dense sand Gravel with stone Medium gravel Clayey sand with gravel Medium gravel Hard sandstone 1< H < 6 60 100 110 140 160 180 200 -

Depth, H (m) 7 < H < 15 250 330 H 15 780 1200

Shear wave velocity for foundation materials

Site Effects (continued)


In alluvial surface layers vibrations are amplified due to multi-reflection effects. The ratio of the amplitude ag at the ground surface to the amplitude at the lower boundary layer (bedrock) ab is given by:

48

ag ab

H cos 2 = + vs

H sin vs
2

1 2

in which is the natural circular frequency of the soil layer and is the wavepropagation impedance:

s vs b vb

where and v are the density and velocity of the surface layer (subscript s) and lower layer (subscript b), respectively.

Site Effects (continued)

49

It is recommended that the ratio between the building and site periods be as distinct from unity as possible. The nature of soil response in earthquakes depends on the amplitude, frequencies and duration of motion. High amplitude motion tends to cause inelasticity in the soil. Long duration shaking increases the susceptibility to liquefaction of saturated and partially saturated soils. When the soil responds elastically, the observed motions at the surface are amplified proportional to the input ground motion. For inelastic response, the soil absorbs large amounts of the energy corresponding to large amplitude ground motions. Large earthquake vibrations travelling through inelastic media will exhibit lower accelerations and large displacements, corresponding to long periods.

Dispersion and Incoherence

50

Dispersion and incoherence may be thought of as the result of the combination of three effects:
i. Wave passage effect: represents the time delay in the arrival of seismic waves on the ground surface at different stations or sites; Extended source effect: number and size of earthquake sources affecting the seismicity at a site may cause delays in the arrival time of waves;

ii.

iii. Ray path effect: caused by reflection and refraction of waves through the soil during their propagation, non-homogeneities of soil layers and other differences in local soil conditions under the various stations.

Sources of local spatial variability of ground motions: wave passage effect (left), extended source effect (middle) and ray path effects (right)

Dispersion and Incoherence (continued)

51

The coherence of two ground motions is a measure of correlation of amplitudes and phase angles at different frequencies. Incoherence is strongly frequency-dependent. Coherence starts to decrease significantly for higher frequencies. Dispersion and incoherence of earthquake ground motions do not generally affect short span structures but they may significantly influence the dynamic (and static) response of long span structures. Significant spatial variability may often occur whenever the large plan dimensions are combined with irregularities in the soil profile along the travel path. For long distances and rather stiff structures totally uncorrelated ground motions with appropriate frequency content should be considered.

Problem 1.4

52

What is the natural period of a layered soil with medium gravel of depth 40m? Is it safe to build a multi-storey building with fundamental period of vibration equal to 1.5 seconds, as that displayed in the figure below, on a site with the above soil type? Is this site more suitable for a particular type of structure shown below?

T=1.5 secs T = 4.5 secs

T=0.2 secs

Vs=780 m/sec

Ts= ?

H=40m

Effects of Earthquakes

53

Comprehensive regional earthquake impact assessment requires an interdisciplinary framework that encompasses the definition of the hazard event, physical damage and social and economic consequences. Such an integrated framework may provide the most credible estimates with associated uncertainty that can stand scientific and political scrutiny. Physical damage should be evaluated for the building stocks, lifeline systems, transportation networks and critical facilities. Short and long term effects should be considered in quantifying social and economic consequences.

Effects of Earthquakes (continued)


Hazard Event Physical Damage
Building Stock

54

Assessment of Impact

Social and Economic Consequences


Short Term
Emergency Shelter, Temporary Housing

Long Term
Housing
Relocation, Displacement

Transportation Systems

Direct Damage, Price Increases, Business Interruption, Supply Disruption

Economic Loss

Fiscal Impacts, Business Failure, Job Loss, Reconstruction

Social Vulnerability

Infrastructure Systems

Casualties, Fatalities, Health Care Disruption

Health

Psychological Distress, Chronic Injury

Critical Facilities

Emergency Supplies, Family Separation

Social Disruption

Family Stress, Neighborhood Disruption

Integrated view of Hazard, Physical, Social and Economic Consequences

Effects of Earthquakes (continued)

55

The fundamental components of earthquake loss assessment are: i. Hazard or exposure; ii. Inventory or assets; iii. Vulnerability or fragility. Hazard or exposure is the description of the earthquake ground motion. Inventory comprises the assets that are subjected to the hazard. Vulnerability or fragility is the sensitivity of the assets to damage from shaking. Seismic risk is the product of hazard and vulnerability for a unit value of assets.

Effects of Earthquakes (continued)

56

Fundamental components for earthquake loss estimations

Effects of Earthquakes (continued)


Earthquake Effects

57

Direct Effects Ground Shaking Ground Cracking Ground Effects Ground Lurching Differential Ground Settlement Soil Liquefaction Lateral Spreading Landslides Rockfalls Structural Effects Vibrations of Structures Falling Objects Structural Damage Structural Collapse

Indirect Effects Ground Effects Other Effects Landslides Tsunamis Seiches Avalanches Rockfalls Floods Fires Toxic Contamination

Direct and indirect earthquake effects

Damage to Buildings and Lifelines


Typical damage to building structures
Masonry & RC Structural element/system Beams Observed damage Structural element/system Beams Steel & Composite Observed damage

58

Shear failure, concrete cover spalling, rebar buckling Cracking, crushing, concrete cover spalling, rebar buckling and pull-out, Columns flexural and shear failure, short column effect Cracking, crushing, rebar buckling Connections and pull-out, shear failure X-shaped cracks, crushing, rebar Structural Walls buckling, overturning, rocking, and Infills sliding Foundations Settlement, rebar pull-out, rocking, sliding, uplifting Soft and weak storeys, excessive residual deformations, distress in diaphragms and connectors, pounding, rocking, uplifting, fall of parapets and brick chimneys

Flange and/or web yielding, local buckling, brittle fracture Flange yielding, local buckling, brittle fracture, splice failure, member buckling Local and member buckling, brittle fracture Yielding, local buckling, brittle fracture, weld cracks, excessive panel deformations, bolt rupture Bolt anchorage rupture, weld cracks and fracture, pull-out, excessive base plate deformations Soft and weak storeys, excessive residual deformations, distress in diaphragms and connectors, pounding, uplifting

Columns

Braces Connections

Foundations

Frames

Frames

Some reconnaissance reports of damage are available at https://www.ideals.uiuc.edu/handle/2142/8798

Damage to Buildings and Lifelines (continued)


Typical types of damage to lifelines
Highways & Railways Bending and shear failure of RC piers Local and overall buckling of steel and composite piers. Brittle fracture of welded components Pounding and unseating at hinge seats and deck supports Cracks, large gaps and/or settlements in pavements of highways

59

Gas & Electric Power Water & Waste Systems Communication Systems Cracks and ruptures in Breakage of pipelines and Damage to electronic the network leakages in the network switching systems Brittle fracture to porcelain components in Sloshing and suction damage Damage to phone lines high-voltage transmission in metal storage tanks stations and substations Damage to switching Elephant foot and shell Damage to telephone systems, cranes and buckling in metal tanks system buildings tanks in power plants Disruptions of electric power supply Cracks and leaks in concrete Malfunctioning of computer basins networks Malfunctioning and/or collapse of transmission towers

Malfunctioning of process Rails bending or rupture Fires and explosions due equipments associated with and train derails to gas leaks ground settlement or rocking

Reconnaissance reports of damage to lifelines are published by the Earthquake Engineering Research Institute (http:\\www.eeri.org)

Effects on the Ground

60

Surface rupture
Rupture of the ground surface may be induced by intense and long shaking as well as direct fault ruptures. These may generate deep cracks and large gaps in the ground. The effects of major fault ruptures can be extreme on structures; buildings can be ripped into parts. Cracks and gaps in the ground may also cause serious damage to transportation systems and underground utility networks. Earthquake-induced ground shaking may cause cracking of the ground surface in soft, saturated soil. Movements of soil or rock masses at right angles to cliffs and steep slopes occur. Structures founded either in part or whole on such masses may experience significant lateral and vertical deformations.

Effects on the Ground (continued)


Surface rupture - Example (Nojima fault on Awaji Island)
Nine kilometres of surface rupture along the Noijma fault on Awaji Island was observed in the 1995 earthquake in Japan.

61

Fault rupture observed in northern Awaji Island during the 1995 Kobe (Japan) earthquake: aerial view with the fault rupture that cuts across the middle of the picture (left) and close-up showing both vertical and horizontal offset of the Nojima fault (right)

Effects on the Ground (continued)


Settlement and uplift
Fault ruptures may cause large vertical movements of the ground. Granular soils are compacted by the ground shaking induced by earthquakes, leading to subsidence. This type of ground movement affects dry, partially saturated and saturated soils with high permeability.

62

Effects of ground settlements and uplift during the 1999 Kocaeli (Turkey) earthquake: flooding (left) and artificial water falls (right)

Effects on the Ground (continued)


Liquefaction

63

Excessive build-up of pore water pressure during earthquakes may lead to the loss of stiffness and strength of soils. The excessive pore water pressure cause ejection of the soil through holes in the ground, thus creating sand boils. The ejection of soil causes loss of support of foundations and thus structures tilt or sink into the ground.

Sand boils during the 1998 Adana-Ceyhan (Turkey) earthquake (left) and the 2001 Bhuj (India) earthquake (right)

Effects on the Ground (continued)

64

Retaining walls may tilt or break from the fluid-like pressure of the liquefied zone. Heavy building structures may tilt due to the loss of bearing strength by underlying soil. Soil liquefaction may cause the floating to the ground surface of pile foundations with low axial loads and underground light-weight storage tanks.

Collapse due to soil liquefaction: settlement and tilting of buildings in the 1964 Niigata (Japan) earthquake (left); soil boils and cracks at pier foundations of Nishinomiya-ko bridge in the 1995 Kobe (Japan) earthquake (right)

Effects on the Ground (continued)


Landslides
Landslides include several types of ground failure and movement. The scale of such landslides on natural slopes can be large enough to devastate entire villages or towns, such as the Huascaran Avalanche triggered by the Per earthquake (1970, Mw = 7.8).

65

Most of the more than 1,000 landslides and rockfalls occurred in the epicentral zone in the Santa Cruz Mountains during the 1989 Loma Prieta earthquake. In the 1994 Northridge earthquake landslides that occurred in Santa Monica, along the Pacific Coast Highway, caused damage to several family houses built on the cliffs overlooking the ocean. Relatively few landslides were triggered by the Hyogo-ken Nanbu earthquake in Japan. This is partly due to the fact that the earthquake occurred during the dry season and the affected area is mainly flat.

Effects on the Ground (continued)

66

In the Kashmir earthquake of October 8, 2005, land-sliding and critical slope stability was a widespread problem that ranged from limited slumping of a superficial nature to complete failure of entire mountain sides.

Effects of a large landslide in Santa Monica in the 1994 Northridge (California) earthquake (left) and satellite view of extensive land-sliding during the 2005 Kashmir (Pakistan) earthquake in the Neela Dandi Mountain (right)

Problem 1.5

67

The 17 August 1999 Kocaeli (Mw=7.4) and 12 November 1999 Dzce (Mw=7.2) earthquakes were the largest natural disasters of the twentieth century in Turkey after the 1939 Erzincan earthquake. These earthquakes caused severe damage and collapse especially of building structures. The figure below shows damage observed in the cities of Adapazari and Izmit (Kocaeli earthquake). Comment on the relationship between the observed damage and the earthquake-induced ground effects illustrated earlier.

Damage observed during the 17 August 1999 Kocaeli in Adapazari (left) and Izmit in Turkey (right)

Human and Financial Losses

68

During the twentieth century, over 1,200 destructive earthquakes occurred worldwide and caused damage estimated at more than $1.0 trillion. Monetary losses from earthquakes are increasing rapidly. Between 1990 and 1999 annual earthquake losses were estimated at $20 billion. The Federal Emergency Management Agency (FEMA) released a study estimating annualized earthquake losses in the USA to be $4.4 billion, with California, Oregon and Washington accounting for $3.3 billion of the total estimated amount. In the USA, 39 out of 50 states (nearly 80%) are at high risk from devastating earthquakes. The Central and Eastern States in the USA now recognise earthquakes as a major threat.

Human and Financial Losses (continued)

69

It is estimated that on average 10,000 people per year were killed by earthquakes between 1900 and 1999. Between 1900 and 2007, deaths due to earthquakes totalled about 1.8 million, caused by 1,250 earthquakes.
300,000
Kansu, China (MS=8.5)

10 8
Magnitude (MS)
Deaths (No.)

300,000
Tangshan, China (MS=7.8)

10 8 6 Magnitude (MS)

250,000
Deaths (No.)

250,000 200,000 150,000 100,000 50,000 0 1978 1989 1990 1998 1994 1983 1971 1976 1980 1987 1992 2001 2005
Deaths Magnitude

200,000 6 150,000 100,000 50,000 0 1906 1923 1939 1950 1915 1948 1960 1963 1964 1970
`

Deaths

Magnitude

4 2 0

4 2 0

Year

Year

Death toll due to earthquakes: 1900-1970 (left) and 1971-2005 (right)

Human and Financial Losses (continued)

70

Several reconnaissance reports have shown that building collapses caused 75% of earthquake fatalities during the last century. Earthquake damage resulting in the collapse of monuments, historical places of worship and stately buildings represents an irreplaceable loss in terms of cultural heritage, while their restoration costs exceed by far the gross national product (GNP) of many affected nations. One of the most severe consequences of earthquakes is the cost of recovery and reconstruction. It is instructive to note, however, that the absolute financial loss is less critical to an economy than the loss as a percentage of the GNP.

Human and Financial Losses (continued)


The business interruption issue has emerged lately as a major concern to industry and hence to communities.

71

Another aspect of the economic impact is the loss of market share which results from interruption to production in industrial facilities and difficulties in re-claiming the share of the market that the affected business previously held.
Country Nicaragua Guatemala Romania Yugoslavia Italy Mexico Greece El Salvador USSR Iran Earthquake Managua Guatemala City Bucharest Montenegro Campania Mexico City Kalamata San Salvador Armenia Manjil Year 1972 1976 1977 1979 1980 1985 1986 1986 1988 1990 Loss ($ bn) 2.0 1.1 0.8 2.2 45.0 5.0 0.8 1.5 17.0 7.2 GNP ($ bn) 5.0 6.1 26.7 22.0 661.8 166.7 40.0 4.8 566.7 100.0 Loss (% GNP) 40.0 18.0 3.0 10.0 6.8 3.0 2.0 31.0 3.0 7.2

Earthquake financial losses

You might also like